A Synergistic Effect Between PG490-88 and Tacrolimus Prolongs Renal Allograft Survival in Monkeys

Abstract

This study was undertaken to determine if PG490-88 and tacrolimus (Tac) act synergistically to prevent renal allograft rejection in monkeys and to explore possible mechanisms of synergy between these agents. MHC-mismatched renal allografts were transplanted into cynomolgus monkeys after bilateral nephrectomy. Recipients were divided into the following groups: (i) no treatment; (ii) PG490-88 (0.03 mg/kg); (iii) Tac (1 mg/kg); (iv) PG490-88 (0.01 mg/kg) + Tac (1 mg/kg) and (v) PG490-88 (0.03 mg/kg) + Tac (1 mg/kg). Through synergy PG490-88 and Tac inhibited anti-CD3/PMA-induced T-cell proliferation and IFN-γ expression in vitro. Tac monotherapy only marginally prolonged survival (27 ± 3.2 days), while the combination of PG490-88 and Tac significantly prolonged graft survival to a median of 99 days (PG490-88 at 0.03 mg) and 38.5 days (PG490-88 at 0.01 mg/kg). Prolonged survival correlated with inhibited IgM production as well as reduced T-cell infiltration, IL-2 protein expression and NF-AT/NF-κB activity. We conclude that PG490-88 and a subtherapeutic dose of Tac significantly prolong renal allograft survival in monkeys through the synergistic inhibition of T-cell activation and a decrease in IFN-γ production and NF-AT/NF-κB activity.

Introduction

Calcineurin inhibitors (CNI), such as cyclosporine A (CsA) and tacrolimus (Tac), have reduced the incidence of acute rejection following organ transplantation. Unfortunately, these agents have not significantly improved long-term graft survival rates due to mortality associated with drug toxicity and chronic rejection (1,2). Therefore, in an attempt to reduce side effects of immunosuppression, one can develop novel immunosuppressive agents that have synergistic effects when combined with CNI. This approach would minimize the dose of CNI required to prevent rejection, thereby reducing drug side effects. These novel immunosuppressive drugs may also have a different mechanism of action and, therefore, may not possess the toxic properties of CNI. These drugs alone, however, do not necessarily result in superior short-term efficacy (3).

Extracts of the Chinese herb Tripterygium Wilfordii Hook F (TWHF) have been used to treat a variety of autoimmune diseases for many years in China (4–6). PG27, an active fraction purified from an extract of this plant, has been used as an immunosuppressive agent to prevent both allograft rejection and graft-vs-host disease (GVHD) in rodents (7–9). PG490 (triptolide) was identified as the most active component of this herb extract, accounting for the immunosuppressive effect of TWHF (10). PG490 has been shown to be a potent immunosuppressant of human Jurkat T cells in vitro, inhibiting T-cell proliferation, IL-2 expression, gamma interferon (IFN-γ) production and NF-κB transcriptional activation (11–13). As well, PG490 in vivo suppresses rejection of cardiac and renal allografts in rodents (8). Because PG490 is not readily water soluble, PG490-88 has been recently developed. This is a novel semi-synthetic, water-soluble derivative that is easier to formulate and administer. PG490-88, a pro-drug of PG490, is converted to PG490 in vivo by hydrolysis of an ester bond (14). The immunosuppressive efficacy of PG490-88 has been reported in GVHD and allograft rejection in both rodent and dog models (8,14,15).

As PG490 inhibits IL-2 gene expression via a different signaling pathway than CsA or Tac, it is logical to expect that PG490-88 may act synergistically with these drugs (11). In fact, the combination of PG27 with CsA has displayed great synergy capable of preventing allograft rejection in rats and inhibiting the production of xenoreactive antibodies in a hamster-to-rat model (8,16). The mechanisms of this synergy, however, have not been explored. Furthermore, PG490/PG27 and/or its pro-drug PG490-88 have not yet been tested in a non-human primate model.

The present study evaluated these questions: (i) Is there synergy between PG490-88 and Tac that can inhibit cellular proliferation and IFN-γ production in vitro? (ii) Can this synergy prevent renal allograft rejection in an in vivo monkey model? (iii) Is synergy associated with in vivo inhibition of IL-2 and NF-AT and NF-κB activity?

Materials and Methods

Animals

Outbred male cynomolgus monkeys were selected as donor and recipient, based on ABO blood group match and genetic non-identity at MHC class II (17). Additionally, the stimulation indices of MLR between donor and recipient pairs prior to transplantation were all higher than 5.

Drugs

a. In vitro cell culture studies:  Stock solution of pure PG490 (Pharmagenesis Inc., Palo Alto, CA), readily bio-available in vitro (not requiring enteric-digestion), and Tac (1000 ng/mL, A.G. Scientific, Inc., San Diego, CA) were dissolved in dimethylsulfoxide (DMSO) and then in RPMI 1640. Both drug stock solutions were further diluted into appropriate working concentrations such that the final concentrations within the microtiter wells were 1, 10, 50 and 100 ng/mL, with DMSO concentrations never exceeding 0.01%.

b. In vivo animal studies:  The pro-drug PG490-88 was provided by Pharmagenesis Inc. and Tac was provided by Fujisawa Healthcare Inc. (Osaka, Japan). Both PG490-88 and Tac were administered daily by gavage.

Cell culture and stimulation conditions

Peripheral blood mononuclear cells were prepared by centrifugation of fresh whole blood from healthy cynomolgus monkeys on a gradient of sodium diatrizoate/Ficoll (Sigma-Aldrich Canada, Oakville, ON, Canada). These isolated cells were stimulated (4 × 10⁶) cells/mL in complete RPMI 1640 medium at various times, depending on the assay to be performed. Stimulations involved incubation with antibody specific to the cynomolgus monkey epsilon unit of CD3 antigen (clone SP34, BD Biosciences, Mississauga, ON, Canada) in the presence of phorbol myristate 13-acetate (PMA; Calbiochem, EMD Biosciences, San Diego, CA). Medium alone served as a negative control. A 1-h drug pre-exposure at 37°C of plated monkey cells to the different concentrations of PG490 and Tac, either alone or in combination, was performed before stimulation.

In vitro assays for T-cell activation and IFN-γ

To assess T-cell activation and effects of the drugs, we measured IFN-γ production and T-cell proliferation simultaneously on triplicate wells for each treatment condition. To measure IFN-γ, supernatants were collected at 72 h and quantitated by a primate-specific ELISA kit (R&D Systems, Hornby, ON, Canada). T-cell proliferation was detected by ³H-thymidine (Amersham, GE Canada, Baie D'Urfe, Quebec, Canada) uptake after 54 h. Proliferation results (cpm) are given as a mean value of ± standard deviation (SD).

Kidney transplantation

Heterotopic kidney allotransplantation was performed on anephric recipients as described previously (18).

Experimental groups

A total of 21 monkeys that received life-supporting kidney transplants were assigned to one of these groups:

Control group: animals untreated (n = 6).

Tac monotherapy (Tac group): recipients given Tac (1 mg/kg/day per os) from day 0 to endpoint (n = 3).

PG490-88 monotherapy (PG 0.03 mg group): recipients given PG490-88 (0.03 mg/kg/day per os) from day 0 to day 30 (n = 4).

PG 0.01 mg + Tac group: recipients given PG490-88 (0.01 mg/kg/day per 05) from day 0 to day 30 and Tac (1 mg/kg/day per os) from day 0 to day 150 (n = 4).

PG 0.03 mg + Tac group: recipients given combination therapy with PG490-88 (0.03 mg/kg/day per os) from day 0 to day 30 and Tac (1 mg/kg/day per os) from day 0 to day 150 (n = 4).

Post-operative monitoring and treatments

After surgery, recipient monkeys were monitored daily for fluid balance, weight, nutritional intake, behavior and general condition. The monkeys were sacrificed when they developed terminal uremia (serum creatinine levels >800 μmol/L). The trough drug concentrations of Tac in whole blood were measured weekly by ELISA technique (19). Using HPLC technique, PG490 trough levels were measured by Alturas Analytics Inc. (Moscow, IN).

Total serum IgG and IgM level assays

Total serum levels of IgG and IgM were measured with the Beckman IMMAGE automated immunonephelometric assay (Beckman, Fullerton, CA) as previously described (20).

Detection of anti-donor specific antibodies

Anti-donor specific IgM and IgG antibodies were detected and measured by flow cytometry using donor lymphocytes as target cells as previously described (21).

Renal histology

Autopsies were performed after the monkeys were sacrificed. Tissue specimens were obtained for histopathological and immunopathological analysis at the time of necropsy or biopsy. The degree of rejection was evaluated blindly by a pathologist (B. G.) according to Banff 97 criterion (22).

Immunohistochemistry

Immunohistochemistry staining was performed using paraffin section. Primary antibodies included rabbit anti-human IgG, rabbit anti-human IgM, rabbit anti-human C3, rabbit anti-human fibrinogen, rabbit anti-human CD3, mouse anti-human CD20 and mouse anti-human CD68 (DakoCytomation, Carpinteria, CA); mouse anti-human CD4 and mouse anti-human CD8 (Novocastra Laboratories Ltd., Newcastle, UK); rabbit anti-human IL-2, mouse anti-human NF-κB and mouse anti-human NF-AT (Santa Cruz Biotechnology, Santa Cruz, CA).

Western blotting

Total kidney protein was extracted from the grafts at the time of sacrifice using lysis buffer (Roche, Laval, Canada). Prepared membranes were probed with the following primary antibodies separately: rabbit anti-human IL-2, mouse anti-human NF-κB and mouse anti-human NF-AT (Santa Cruz Biotechnology). Quantitative analyses of the Western blots were performed as previously described (21).

Quantitative cell counts in renal grafts

Five fields of prepared graft biopsies were randomly selected and digital pictures were taken from each immunohistochemical staining section of CD3, IL-2, NF-κB and NF-AT under 250-fold magnification. Cell counting was done using the Image-Pro Plus program (SPSS Inc., Chicago, IL) as previously described (23).

Statistics

Graft survival data are expressed as median with a range (± SE). Comparisons of parametric data were made with SigmaStat software using one-way analysis of variance and non-parametric data were compared using a Mann-Whitney Rank Sum test or Log Rank test. P values less than 0.05 were considered statistically significant. To confirm synergy of in vitro inhibition by PG490 and Tac, a median-effect analysis was used (24). The interaction between the two drugs is assessed by the combination index (CI). CI < 1, =1, and >1 indicates synergism, additive and antagonism, respectively.

Results

PG490 and Tac synergistically inhibit cell proliferation in vitro

Although it has been previously shown that either PG490 or Tac alone inhibits lymphocyte proliferation in vitro (11,25), it has not been investigated if these agents can act synergistically to inhibit cellular proliferation in vitro. Figure 1A shows that anti-CD3 antibody in the presence of PMA (anti-CD3/PMA) stimulated cell proliferation. Neither PG490 nor Tac alone at concentrations of 1 and 10 ng/mL, respectively, were capable of significantly inhibiting cell proliferation. The combination of both agents at these concentrations, however, markedly inhibited T-cell proliferation by 50% (p < 0.01 vs control group). This synergy between PG490 and Tac on in vitro T-cell proliferation was observed at different concentrations of the two drugs and was dose-dependent (C = 0.463).

Figure 1. PG490 and Tac synergistically inhibit anti-CD3/PMA-stimulated T-cell proliferation (A) and IFN-γ production (B) in vitro. Cell proliferation was monitored in vitro by ³H-thymidine uptake at 72 h. IFN-γ was quantitated from supernatants collected 72 h after stimulation. There is evidence of significant differences in T-cell proliferation and IFN-γ production between the presence of PG490 or Tac alone and the combination of both agents at 1, 10 and 100 ng/mL (p < 0.01).

PG490 and Tac synergistically inhibit IFN-γ production in vitro

This phenomenon of drug synergy was also apparent when IFN-γ production of T-cells was examined (Figure 1B). Low levels of PG490 were capable of diminishing the amount of IFN-γ expressed by the anti-CD3/PMA activated T cells, with 50% inhibition concentration (IC₅₀) of PG490 being about 2 ng/mL, and complete inhibition was achieved at 32 ng/mL. In contrast, measuring the effect of Tac alone on IFN-γ production revealed an IC₅₀ of Tac at 50 ng/mL with no further dose-dependent inhibition of cytokine production, even at levels of 1000 ng/mL. Of particular interest were the synergistic effects of the two drugs on T cells activated with anti-CD3/PMA. Utilizing Tac at in vivo trough levels of 10 ng/mL in combination with PG490 at 1 ng/mL had a significant impact on in vitro IFN-γ production. This finding indicates the synergy of the two drugs, especially as Tac alone at 10 ng/mL to 500 ng/mL had a limited maximum inhibitory effect of 50% on T-cell cytokine production, while upon addition of PG490 cytokine production was significantly further inhibited. This synergy between PG490 and Tac on in vitro inhibition of IFN-γ production was observed at different concentrations of the two drugs and was dose-dependent (C = 0.177).

PG490-88 and Tac synergistically prolong renal graft survival

Next, we investigated whether the drugs acted synergistically in preventing allograft rejection and prolonging survival in a monkey kidney allograft model. To test this hypothesis, PG490-88 was combined with a subtherapeutic dose of Tac. Table 1 shows individual survival data and cause of death. Untreated allografts were rapidly rejected within 9 days. PG490-88 monotherapy at a dose of 0.03 mg/kg/day failed to significantly prolong allograft survival, except for one animal that survived for 28 days. Tac monotherapy at a dose of 1 mg/kg/day moderately prolonged the median survival to 27 days. In contrast, the combination of PG490-88 (at a dose of either 0.03 mg/kg or 0.01 mg/kg) and the same subtherapeutic dose of Tac significantly prolonged renal allograft survival to median survival of 99 days and 38.5 days, respectively (p < 0.01 vs control group; p < 0.05, PG 0.03 mg + Tac group vs Tac alone group). Notably, 50% of animals treated with PG490-88 0.03 mg/kg and Tac 1 mg/kg survived longer than 150 days. Another animal treated with PG490-88 0.01 mg/kg and Tac 1 mg/kg also survived for more than 150 days. One of the animals treated with combination therapy was sacrificed for pathology studies on day 154 with almost normal renal function (Figure 2). Tac was discontinued on day 150 in two other 'combination' treated animals and their grafts finally rejected on day 182 and day 246. Figure 3 shows Tac trough levels in the Tac alone group and the two combination groups. There were no statistical differences among the groups (p = NS, analysis of variance (ANOVA)). Similarly, there was no statistical difference in PG490 trough levels between PG490-88 treated alone and combination groups (the mean level of PG490: 0.2 ± 0.03 ng/mL for PG490-88 alone group vs 0.3 ± 0.05 ng/mL for PG490-88 ± Tac, p = NS, ANOVA).

Table 1.  Individual survival, cause of death and pathology

Animal ID	Survival (days)	Median survival ± SE (days)	Cause of death	Banff 97 type
*Animal developed severe diarrhea due to amoebiasis on day 8; Tac was withdrawn for one day; the dose was reduced afterward as a result of seizures. The animal was euthanized on day 17 owing to poor general condition. ^†Tac was discontinued from day 150. ^‡The animal was sacrificed for pathology studies. ^ap < 0.05 vs control group (Mann-Whitney Rank Sum test). ^bp < 0.01 vs control group (Mann-Whitney Rank Sum test). ^cp < 0.03 vs either Tac alone group or PG 0.03 mg group (Log Rank test).
Control group
12-00	5		Rejection	III
13-00	6		Rejection	III
15-00	7	7 ± 0.6	Rejection	III
17-00	8		Rejection	III
35-02	9		Rejection	III
37-02	7		Rejection	IIB
PG 0.03 mg group
22-02	1		Respiratory failure	Normal
26-02	8	8 ± 6.8	Rejection	IIA
28-02	28		Rejection	IIA
46-03	7		Rejection	IIA
Tac alone group
38-02	23		Rejection	IIA
39-02	27	27 ± 3.2^a	Rejection	IIA
47-03	34		Rejection	IIA
PG 0.01 mg + Tac
33-02	32		Rejection	IA
44-03	44	38.5 ± 37.1^b	Rejection	IB
51-03	26		Rejection	IB
54-03	182^†		Rejection	IIA
PG 0.03 mg + Tac
29-02	17*		Sacrificed*	IA
45-02	44		Rejection	IB
30-02	154^‡	99 ± 52.7^b,c	Sacrificed^‡	Borderline
52-03	251^†		Rejection	Chronic I

Figure 2. Serum creatinine levels of renal allograft recipients in different groups. Arrows indicate the time at which Tac was discontinued on post-operative day (POD) 150.

Figure 3. Serum tacrolimus trough levels in different groups. p = NS, ANOVA.

The combination of Tac and PG490-88 improves renal function

Figure 2 illustrates renal function as measured by serum creatinine levels. Most recipients treated with Tac or PG490-88 alone developed uremia within 30 days. In contrast, all recipients except one treated with the higher dose PG490-88 combination exhibited normal renal function for 30 days. One recipient in this group was sacrificed on day 17 owing to poor general condition. Only one animal from the lower dose PG490-88 combination group developed uremia within 30 days and the other three animals had slightly increased serum creatinine levels during this period. When PG490-88 was discontinued on day 30, the serum creatinine levels were moderately elevated and the animals developed uremia within 1–3 months after withdrawal of Tac.

PG490-88 alone or in combination with Tac decreases circulating IgM levels and donor-specific IgM levels after allotransplantation

Circulating total IgM levels were markedly elevated after transplantation in all animals treated with both Tac monotherapy or in combination with the lower dose of PG490-88. In contrast, circulating IgM levels were significantly decreased in the PG490-88 monotherapy and higher dose PG490-88/Tac combination groups. Figure 4A demonstrates that the mean total IgM serum level in animals, treated with 0.03 mg/kg of PG490-88 and Tac, was significantly lower than those of the Tac alone group or the lower dose PG490-88 combination group at days 14, 21 and 28 (p < 0.05). These findings were further confirmed by measuring donor-specific IgM antibodies in circulating blood. Figure 4B shows that PG490-88 0.03 mg/kg alone or combined with Tac 1 mg/kg completely inhibited the elevation of inducible donor-specific IgM levels in circulation, while Tac alone or combined with PG 0.01 mg/kg failed to inhibit inducible IgM production.

Figure 4. PG490-88 at a dose of 0.03 mg/kg/day alone or combined with Tac inhibits IgM production following transplantation. (A) The mean total serum IgM levels in different groups. The total serum IgM levels in recipients treated with combination of PG0.03 mg/kg and Tac 1 mg/kg were significantly lower than those treated with Tac alone or combined with PG 0.01 mg/kg (*p < 0.01 vs combination of PG490-88 0.01 mg/kg and Tac; ^#p < 0.05 vs Tac treated alone group). (B) Donor specific IgM serum levels on POD 21 measured by FACS in a representative animal. Red line represents baseline of IgM and shaded curve represents IgM levels on day 21. PG490-88 0.03 mg/kg alone or combined with Tac 1 mg/kg completely inhibited donor-specific IgM production, while Tac alone or combined with PG 0.01 mg/kg failed to inhibit IgM production. (C) Immunohistochemical staining of IgM in renal allografts with different treatment at necropsy.

These findings were further supported by immunohistochemistry studies. As shown in Figure 4C, intragraft IgM deposition was significantly reduced in both the PG490-88 monotherapy group and the combination of PG 0.03 mg/kg and Tac group, while Tac alone or combined with PG 0.01 mg/kg failed to inhibit IgM deposition. This observation indicates that PG490-88 at a dose of 0.03 mg/kg has a unique capacity to inhibit IgM production. There were no significant differences in total serum IgG and donor specific IgG during the first month among these groups (data not shown).

The combination of PG490-88 and Tac attenuates renal allograft rejection

The Banff score was used to evaluate the pathological changes at the end point, and the score for each graft is presented in the supplemental data section. Untreated allografts developed advanced cellular and vascular rejection with a median score of III. Grafts treated with PG490-88 or Tac alone developed moderate rejection with a median score of II. In contrast, grafts in the combination groups showed minimal rejection with a median score of I. The renal graft from a monkey treated with combination therapy, and sacrificed on day 154 for pathology studies, revealed near normal histology. Biopsies at 150 days from the other two animals treated with combination therapy showed minimal evidence of rejection or only early signs of chronic rejection. These data indicate that the combination of PG490-88 and Tac attenuated renal allograft rejection. These two grafts demonstrated a pattern of chronic rejection on days 182 and 251 after withdrawal of Tac on day 150.

The combination of PG490-88 with Tac significantly inhibits T-cell infiltration

Figure 5 shows that the combination of PG490-88 and Tac significantly inhibited T-cell infiltration in the graft, using quantitative morphological analysis of CD3 positive cells. In the control group, the CD3 positive cells were widespread in the renal parenchyma with accumulation around the artery, vein and in the glomeruli (Figure 5A-a), while the allografts in the Tac and PG490-88 treated alone groups had less CD3 positive cellular infiltration in interstitium, and showed focal accumulation around the artery and beside glomeruli. (Figure 5A-b, A-c). Notably, renal allografts in the PG490-88 and Tac combination groups only revealed a small patch of CD3 positive cells surrounding a small artery (Figure 5A-d, A-e). Quantitative morphological analysis (Figure 5B, left first panel) shows that the number of CD3 positive cells from renal allografts, treated with PG490-88 and/or Tac, was significantly reduced compared to untreated controls (p < 0.001). There was a further reduction of CD3 positive cells in allografts treated with combination therapy (p < 0.001 vs control). There was a significant difference in CD3 positive cells between PG490-88/Tac monotherapy groups and combination groups (p < 0.01).

Figure 5. (A) CD3, IL-2, NF-kB and NF-AT immunohistochemical staining of a representative renal graft at necropsy from different groups: an untreated control (15-00, POD 7, Banff score III, first row) showing massive positive cells stained with CD3 (a), IL-2 (f), NF-kB (k) and NF-AT (p); a PG 0.03 mg/kg/d treated allograft (26-02, POD 8, IIA, second row) showing moderate positive cell infiltration stained with CD3 (b), IL-2 (g), NF-kB (i) and NF-AT (q); a Tac 1 mg/kg/day treated allograft (39-02, POD 27, IIA) showing moderate positive cell infiltration stained with CD3 (c), IL-2 (h), NF-kB (m) and NF-AT (r) ; a PG490-88 0.01 mg/kg/day + Tac 1 mg/kg/day treated allograft (44-03, POD 44, IB, fourth row) showing a minimal number of positive cells stained with CD3 (d), IL-2 (i), NF-kB (n) and NF-AT (s); and a PG490-88 0.03 mg/kg/day + Tac 1 mg/kg/day treated allograft (30-02, POD 153, electively explanted, Banff score borderline, fifth row) showing a minimal number of positive cells stained with CD3 (e), IL-2 (j), NF-kB (o) and NF-AT (t). (B) Quantitative analysis of CD3, IL-2, NF-kB and NF-AT. The data represent the mean value of four animals and the arrow bar represents SD of the mean. CD3 positive T-cell numbers in the graft counted by Image-Pro Plus program (first left column). IL-2 (second left column), NF-kB (third left column) and NF-AT (forth left column) were measured by semi-quantitative Western blotting. The number of CD3 positive cells infiltrating was significantly reduced in Tac or PG490-88 monotherapy groups (b and c) (p < 0.001 vs controls, a). The number of CD3 positive cells was further reduced in combination of PG490-88 and Tac groups at PG490-88 dose of 0.01 mg/kg (d) or 0.03 mg/kg (e) (p < 0.001 vs Tac/PG490-88 monotherapy). IL-2 protein expression was moderately suppressed by PG490-88 or Tac monotherapy (g and h, p < 0.05 vs untreated control, f). There was significant difference of IL-2 expression between combination therapy groups (i and j) and monotherapy by PG490-88/Tac (p < 0.01). NF-AT protein expression was dramatically inhibited by the combination of PG490-88 and Tac treatment (n and o, p < 0.001 vs control, k), while NF-AT expression was only moderately suppressed by PG490-88 or Tac alone treatment (i and m, p < 0.001 vs untreated control, k). NF-κB protein expression was strongly inhibited by combination of PG490-88 and Tac treatment (s and t, p < 0.001 vs control, p), while NF-κB expression was only moderately suppressed by PG490-88 or Tac alone treatment (q and r, p < 0.001 vs untreated control, p).

The combination of PG490-88 with Tac markedly inhibits IL-2 protein expression in renal allografts

To study the potential mechanisms of synergy between PG490-88 and Tac in preventing renal allograft rejection in this model, we investigated IL-2 protein expression in the renal grafts at the endpoint using immunohistochemical staining and Western blotting. Immunohistochemical staining showed that most infiltrated T cells were IL-2 positive in the untreated control group (Figure 5A-f). IL-2 positive cell numbers were reduced in the PG490-88 and Tac treatment alone groups (Figure 5A-g, A-h). PG490-88 monotherapy determined slightly stronger inhibition than Tac monotherapy, but there was no significant difference between the two groups (p > 0.05). Renal allografts with combination therapy demonstrated a minimum number of IL-2- positive cells, which were surrounding arteries. IL-2 positive cells were rarely seen in the interstitium (Figure 5A-i, A-j). Western blotting confirmed these results. As shown in Figure 5B (second left panel), IL-2 protein expression was moderately suppressed in PG490-88 and Tac monotherapy groups (p < 0.05 vs control) while IL-2 protein expression was markedly suppressed in renal allografts treated with combination therapy (p < 0.001 vs control).

The combination of PG490-88 with Tac markedly inhibits NF-AT and NF-κB transcriptional activation in the graft

Previous in vitro studies showed that either PG490-88 or Tac inhibited IL-2 expression from T cells at the level of NF-AT and NF-κβ activation. It remains unknown if the synergy that delays rejection is associated with inhibition of NF-AT and NF-κB transcriptional activation in the graft. We compared NF-AT and NF-κB expression from renal allografts in different treatment groups at the endpoints using immunohistochemical staining and Western blotting. Similar to the pattern seen with IL-2 positive cells, the NF-AT and NF-κB positive cells were widely distributed, forming a heap of cells around arteries, veins and glomeruli in untreated allografts (Figure 5A-k, A-p). PG490-88 or Tac monotherapy had moderate inhibition of NF-κB and NF-AT (Figure 5A-i, A-q, A-m, A-r), while combination therapy markedly inhibited both NF-κB and NF-AT expression (Figure 5A-n, A-s, A-o, A-t). Western blotting substantiated these findings (Figure 5B third and fourth left panel). Either PG490-88 or Tac partially suppressed NF-κB and NF-AT expression (p < 0.05 vs control), while combination therapy markedly inhibited both NF-κB and NF-AT expression (p < 0.01). There was a significant difference in NF-AT/NF-κB expression between PG490-88/Tac monotherapy and combination therapy groups (p < 0.01).

Potential side effects of oral PG490-88

Oral administration of PG490-88 at a dose of 0.03 mg/kg or 0.01 mg/kg was well tolerated by the monkeys with no vomiting or diarrhea. There was no significant difference in weight change among recipients of PG490-88 monotherapy or combination therapy, controls and Tac monotherapy group (data not shown). There were no differences in liver function (ALT, AST levels), white blood cell and red blood cell counts among all groups (data not shown).

Pathology studies at necropsy showed that there was no evidence of toxicity in the liver, heart, lungs, pancreas or gastrointestinal system in animals treated with PG490-88 alone and/or Tac.

Discussion

This is the first report to demonstrate the synergistic activity from the combination of PG490-88 and Tac in preventing allograft rejection in a functional kidney transplant model of non-human primates. One month of PG490-88 treatment at a dose of 0.03 mg/kg/day, combined with a subtherapeutic dose of Tac, allowed 50% of monkeys to survive more than 150 days, while the same dose of Tac monotherapy only marginally prolonged survival to 27 days. Combination therapy significantly improved renal graft function and attenuated rejection. PG490-88 alone or combined with Tac effectively inhibited IgM production following transplantation, while Tac alone failed to inhibit IgM production. Our in vitro data show that the combination of these agents had a synergistic effect of inhibiting T-cell proliferation and IFN-γ production. In fact, utilizing Tac at an average in vitro level of 10 ng/mL in combination with pure PG490 at 1 ng/mL had a significant impact on T-cell proliferation and IFN-γ production, indicating the synergy of the two drugs. This was especially evident in light of the fact that Tac alone at this same level had little or no effect on proliferation and cytokine production by the T cells. These findings were further confirmed by tests using in vivo immunochemistry and Western blotting. We found that the synergy of both agents was associated with a marked inhibition of IL-2, NF-AT and NF-κB in the grafts. These encouraging results suggest that PG490-88 may be of value as a potent adjunct agent in combination with conventional immunosuppressive drugs in future clinical transplantation.

Another important finding from this study is the unique inhibition of IgM by PG490-88. Elevation of IgM has been reported in inadequately immunosuppressed recipients following both experimental and clinical transplantation (26,27). In this study, PG490-88 monotherapy or combined with Tac markedly inhibited circulating total IgM and donor-specific IgM levels and prevented intragraft IgM deposition following transplantation, whereas Tac alone failed to inhibit IgM. These results support the previous observation in a hamster-to-rat cardiac xenograft model that the combination of PG27 and CsA, not CsA alone, significantly inhibited IgM xenoantibody production (16). We also previously reported that CsA failed to inhibit IgM production in a rat-to-mouse xenograft model (28). The precise mechanisms of inhibiting IgM by PG490-88 is unknown, and we speculate that it might result from the unique mechanism of PG490-88, which involves transcriptional inhibition of NF-κB signaling in the nucleus at a novel, distal site in the signaling pathway (11). In addition, triptolide has been reported as a potent suppressant of C3, CD40 and B7 expression in activated human proximal tubular epithelial cells (29). The inhibition of complement by PG490-88 has been seen in our dog allograft model (15).

Until recently, the role of acute humoral rejection (AHR) has been underestimated. With the development of sensitive FACS technologies to detect circulating antibodies and using C4d as a marker for diagnosis of AHR, this type of rejection has been recognized as a major challenge following transplantation (30). AHR is typically unresponsive to conventional immunosuppressive agents including Tac. Although the role of IgM in AHR is not well defined, we believe that inhibition of IgM and complement by PG490-88 may have a potential value to prevent AHR in future clinical transplantation.

This study has also demonstrated that PG490-88 oral administration at a dose of either 0.03 mg/kg or 0.01 mg/kg did not show any significant side effects in non-human primates. Either dosage level of PG490-88 did not cause renal or pancreatic toxicity in monkeys. Whether increasing the dose or treatment duration of PG490-88 would improve the efficacy or result in toxic side effects needs to be further investigated. Because there is no long-term toxicity data about PG490-88, this agent was only used for 1 month in this study. Most rejection episodes occurred after withdrawal of this agent in the combination groups. It is warranted to test whether continuous use of PG490-88 in combination with a subtherapeutic dose of Tac would further improve the efficacy of this regimen.

We conclude that this novel combination regimen significantly prolongs renal allograft survival in monkeys through synergistic inhibition of IL-2 and IFN-γ expression as well as NF-AT and NF-κB transcriptional activity. PG490-88 also has a unique capacity to inhibit IgM production. These encouraging results suggest that PG490-88 may be a valuable novel therapy as an adjunct agent with conventional immunosuppressive drugs in future clinical transplantation.

Acknowledgments

The authors thank Tamie Fulford, Kim Lansbergen and Rachel Daniels for excellent post-operative studies and animal care; Dr. Ian Welch, Kim Thomaes and Lynn Denning for excellent veterinarian and technical assistance; Dr. Gill Strejan for reviewing the manuscript, Cate Abbott for editorial assistance and Sharon Mutch for secretarial support. This work was supported by Fujisawa Pharmaceutical Co. Ltd., Japan, the Multi-Organ Transplant Program, London Health Sciences Centre and NIH grant U19 AI51731-01.